Description
CodeFuse-ChatBot is an open-source AI intelligent assistant developed by the Ant Group's CodeFuse team, dedicated to simplifying and optimizing various stages of the software development lifecycle. The project integrates a Multi-Agent framework for collaborative scheduling, along with rich toolkits, code repositories, knowledge bases, and a sandbox environment. This enables Large Language Models (LLMs) to effectively execute and handle complex tasks within the DevOps domain.
The core objective of CodeFuse-ChatBot is to build an AI assistant for the entire software development lifecycle, encompassing design, coding, testing, deployment, and operations. It aims to shift traditional development and operations models, which involve extensive manual data querying and platform operations, towards an intelligent development and operations model powered by LLM Q&A, thereby transforming user development and operations habits.
Key differentiating technologies and features include a robust intelligent scheduling core that supports multi-mode one-click configuration, simplifying operational workflows. It offers warehouse-level in-depth code understanding and project file-level code writing and generation, significantly boosting development efficiency. The system enhances document analysis through the integration of document knowledge bases and knowledge graphs, providing deeper support via retrieval and reasoning.
CodeFuse-ChatBot also provides a specialized knowledge base for the DevOps domain, supporting one-click self-building of domain-specific knowledge bases for convenience and practicality. It ensures compatibility with smaller models tailored for the DevOps domain, promoting the integration of the technical ecosystem. The project supports offline private deployment based on open-source LLM and Embedding models, and also offers OpenAI API integration.
The project is built upon Langchain-Chatchat and codebox-api, with contributions from a core R&D team focused on AIOps + NLP research. The goal is to enable widespread contributions of high-quality development and operations documentation to collectively improve this solution, aiming to achieve the vision of 'making development easier for everyone'.
CodeFuse-ChatBot's Core Features
Multi-Agent Framework for intelligent scheduling
DevOps Toolkits integration
Code and Document Repository RAG (Retrieval Augmented Generation)
Warehouse-level code analysis
Project file-level code generation
Document knowledge base and knowledge graph integration
Domain-specific knowledge base for DevOps
Support for open-source LLM and Embedding models
Offline private deployment capability
OpenAI API support
Sandbox environment for code execution
Multi-source web crawler for information gathering
Getting Started with CodeFuse-ChatBot
Clone the repository
Install dependencies
Configure server_config.py
Start the service using bash start.sh
Configure settings in the web page
Launch the conversation service
CodeFuse-ChatBot's Use Cases
- Code Generation
- Code Understanding
- Document Analysis
- DevOps Automation
- Intelligent Q&A
- Private Knowledge Augmentation
